Microstructural characterization of Ni-201 weld cladding onto 304 stainless steel

Accepted Manuscript Microstructural characterization of Ni-201 weld cladding onto 304 stainless steel

Xianwu Shi, Kun Yu, Li Jiang, Chaowen Li, Zhijun Li, Xingtai Zhou PII: DOI: Reference:

S0257-8972(17)31160-X doi:10.1016/j.surfcoat.2017.11.023 SCT 22871

To appear in:

Surface & Coatings Technology

Received date: Revised date: Accepted date:

21 July 2017 6 November 2017 8 November 2017

Please cite this article as: Xianwu Shi, Kun Yu, Li Jiang, Chaowen Li, Zhijun Li, Xingtai Zhou , Microstructural characterization of Ni-201 weld cladding onto 304 stainless steel. The address for the corresponding author was captured as affiliation for all authors. Please check if appropriate. Sct(2017), doi:10.1016/j.surfcoat.2017.11.023

This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT Microstructural characterization of Ni-201 weld cladding onto 304 stainless steel XianwuShi1,2, Kun Yu1,2, Li Jiang1, Chaowen Li1,Zhijun Li1*,Xingtai Zhou1 1

Center for Thorium Molten Salt Reactor System,Shanghai Institute of Applied Physics, Chinese

Academy of Sciences, Shanghai 201800, China 2

University of Chinese Academy of Sciences, Beijing 100049, China

Chinese

Academy

of

Sciences,

Shanghai

201800,

China;

Email:

RI

Physics,

PT

 Corresponding author : Thorium Molten Salts Reactor Center, Shanghai Institute of Applied

[email protected] (Li Zhijun); Tel./Fax: +86-2139194767

SC

Abstract

Ni-201(ERNi-1), a cladding layer which is resistant to the corrosion of molten salts, was

NU

deposited on 304SS substrate by Gas Tungsten Arc Welding (GTAW). Microstructure characterization showed that the cladding interface was obviously divided into three zones

MA

from microstructures along depth, weld metal (WM), unmixed zone (UZ) and heat affected zone (HAZ). Element distribution presented that elements Ti, N, Si were segregated into

D

interdendritic region of WM and formed the TiN precipitates during the solidification. The

PT E

element distribution of UZ was similar to that of HAZ. A large quantity of vermicular δ-ferrite phases were precipitated in laminar UZ. Grain coarsening in HAZ was evident compared with the grains in base metal, and no precipitates were found in HAZ. The hardness of cladding

CE

layers was decreased from interface to surface. The cladding layer exhibited excellent corrosion resistance to molten FLiNaK salts.

AC

Keywords: Gas Tungsten Arc Welding (GTAW); Microstructure; Interface; Precipitates; δ-ferrite; Hardness; Corrosion resistance 1. Introduction

Molten salt reactor (MSR) is the only reactor made use of high temperature molten salts as fuels and coolants among the fourth generation nuclear energy systems. MSR possesses a large quantity of advantages such as uncomparable safety, simple and flexible fuel cycle in operation, better fission fuels utilization, 40%-50% power generation efficiency, hydrogen by product at high temperature and so on [1]. In recent years, the development of MSR has attracted more and more attention from the international nuclear communities [2-4]. 1

ACCEPTED MANUSCRIPT In the fluoride molten salt reactor, the structural materials must possess high-temperature strength and chemical compatibility with the molten fluoride salt as well as good irradiation resistance. A solid-solution strengthened Nickel-based alloy (Hastelloy N and GH3535, in the same category) has been developed in USA and China [4]. This kind of alloy has excellent high temperature stability and good corrosion resistance to molten fluoride salts [5-8]. However, it has also some problems as follows: (1) as a Nickel-based alloy, its cost is much

PT

higher than that of stainless steel, (2) the large-size plates and forgings of alloy are much difficult to obtain due to the limitation of melt method and processing facility abilities, (3) up

RI

to date, this alloy cannot be used in a commercial reactor because its high temperature

SC

performance data are insufficient in ASME Boiler and Pressure Vessel Code [9]. Weld overlay cladding is used for corrosion resistance or wear resistance and employed to

NU

repair a worn parts in service or follow a manufacturing anomaly, such as turbine blades in a power plant [10, 11]. In chemical application industry, some equipments are made from

MA

welding overlay super stainless steel claddings onto carbon steels to prevent chemical liquid corrosion [12]. In the nuclear power industry, weld overlay claddings made of stainless steel are used on the inner surfaces of pressurized water reactor vessels (PRV) as protective barriers

D

against corrosion [13]. At the safe-end joints of PRV, Nickel-based alloys are overlaid onto

PT E

low-alloy steels, and then welded with stainless steels to prevent the carbon diffusion and ensure the reliability of welded joints [14].

CE

Weld overlay cladding that developed during fusion welding of dissimilar alloys will achieve a chemical composition intermediate region between the base metal and filler metal.

AC

The chemical composition will, in turn, have a significant influence on the microstructure, mechanical properties and corrosion resistance of weld metal. Many researchers focused on the evaluation of microstructure and performance properties of weld overlay cladding. Dupont et al. [15] found that the dilution of element Fe expanded the solidification temperature range and affected the length of cracks between AL-6XN super austenitic stainless steel and IN625 Nickel-based alloy. A. Mortezaie et al. [16] pointed that solidification cracking was caused by the formation of brittle Cr23C6 phases along the grain boundaries in dissimilar joint between Inconel 718 Nickel-based superalloy and 310S austenitic stainless steel. Therefore, it is necessary to study the microstructures of the interface between the base metal and weld 2

ACCEPTED MANUSCRIPT overlay cladding. In order to reduce the cost of the structure materials used in fluoride molten salt reactor, 304 stainless steel is chosen as the base metal because of its economical price and complete high temperature data[17], and Ni-201(ERNi-1) is chosen as the weld overlay cladding because of its excellent corrosion resistance to molten fluorite salts[18- 23]. In this work, ERNi-1 was deposited on the surfaces of 304 stainless steel substrate by Gas

resistance of cladding layer were investigated and discussed.

RI

2. Experiment

PT

Tungsten Arc Welding (GTAW). The microstructures, hardness, molten-salts corrosion

SC

The base metal was 304 stainless steel and the filler metal was AWS ERNi-1, their chemical compositions were given in Table 1, respectively. The base metal plate was cleaned

NU

with a grinding wheel to remove the oxide and impurities and then wiped by acetone solution. The welding procedure was performed in multiple passes by GTAW process with a cold wire

MA

automatic feeding system. The welding parameters given in Table 2 had been optimized according to weld appearance quality and nondestructive testing results. The heat input (Q) was calculated by Q=ηEI/V(η=0.7) [24], where η was thermal efficiency, E was the welding

D

voltage, I was the current and V was welding speed.

PT E

For metallographic examinations, transverse section samples were cut by an electric spark cutting machine, followed by grinding with SiC papers from 400 grits to 2000 grits, and then

CE

followed by polishing with a 0.05μm alumina paste for 2 min. The sample was etched in a mixed acid (5gFeCl3+50mlHCl+100mlH2O) for 304 stainless steel and then etched in

AC

another solution (100mlCH3COOH+38mlHNO3+10mlH2O) for cladding. Another sample was electrolytic etched in 10% ammonium persulphate solution for unmixed zone (UZ). The sample for EBSD analysis was electrolytically etched in a solution (20%HClO4+CH3COOH). The microstructure was investigated using OM, SEM, and EBSD. And the element analysis was carried out by EDS and EPMA. TEM samples were prepared following a standard procedure described in [25] which includes dimple polishing both sides of the sample followed by Ar+ ion milling in a Gatan precision ion polishing system (PIPS II) with liquid N2 cooling. The hardness tests were measured with indentations at intervals of 0.2mm under a load of 3

ACCEPTED MANUSCRIPT 200gf and a dwell time of 15s. The cladding layers consisted of three weld metal layers. The samples used for static corrosion tests were cut from the top of cladding layers with sizes of 20mm×10mm×2mm. The FLiNaK salts (LiF-NaF-KF: 46.5-11.5-42mol.%) were used for the tests. Fig.1 showed the schematic diagram of the corrosion test [8, 18, 20]. The container was high-purity graphite crucible and the samples were suspended with ERNi-1 wire due to the low galvanic corrosion

PT

between the wire and the tested sample [26]. Stainless steel shield covers with lids were chosen to reduce oxidation of the graphite crucibles at high temperature. All the experimental

RI

setup was operated in a glove box under argon atmosphere. The samples were immersed into

SC

the 250g FLiNaK salts and corroded at 650°C for 200h. The samples after corrosion were cleaned by 1 mol/L Al (NO3)3 solution and deionized water. The microstructures and element

NU

distribution of corroded samples were examined in accordance with previous procedures. 3. Results and discussion

MA

3.1 Microstructure and elements distribution of cladding interface Fig.2a showed the microstructure of cladding interface by OM. It presented that the regions after welding can be divided into three zones which were named weld metal (WM), unmixed

D

zone (UZ) and heat affected zone (HAZ) respectively. There were no defects such as pores,

PT E

cracks formed in the three zones. The interface exhibited an excellent metallurgical bonding performance. The microstructure of WM consisted of fully austenitic grains with a columnar

CE

dendritic morphology. Quantities of vermicular precipitates (black phases in Fig. 2a) were obvious in UZ with a width of approximately 100µm. It was noticed that the width of UZ

AC

varied along the interface which was attributed to the different heat inputs as a result of the swing of arc during welding. The arc stayed at both sides of the welding beads, which led the heat input at the sides higher than that in middle and eventually formed the unequal width of UZ. In addition, the heat cycle at the overlapping area of two beads influenced the width of UZ too. The microstructure of HAZ consisted of equiaxed austenite grains and a small amount of annealing twinning. Moreover, some residual high-temperature δ-ferrite in the form of stringers existed in the grains. Fig. 2b showed the interface between UZ and WM. It can be seen that the grains exhibited epitaxial growth from UZ to WM. Although the ERNi-1 filler metal and 304SS have great differences in composition, they exhibited the same crystal 4

ACCEPTED MANUSCRIPT structure and nucleation of solid weld metal by arranging atoms from the liquid metal on the substrate grains with special crystallographic orientations. This was the reason for the epitaxial growth observed at the interface [27]. Fig. 3 showed the distribution of elements from HAZ to WM by EDS. It was found that the chemical composition was similar to that of HAZ. However, the element fluctuation in UZ was more evident than that in HAZ because of the precipitates in UZ. There was a

PT

composition transition region with a width of 25μm in WM. It can be seen that the contents of Cr and Fe decreased sharply indicating the dilution levels of Cr and Fe are high. According to

RI

the EDS analysis, an obvious peak of Ti was found in WM which suggested the presence of

SC

the Ti-rich precipitates. The composition of WM was qualitatively analyzed by EDS as shown in Table 3. The chemical composition of WM near the interface especially the content of Ni,

NU

Fe and Cr is much more different from that of ERNi-1 shown in Table 1as a result of higher dilution of dissimilar welding.

MA

3.2 Microstructure of the cladding (WM)

Fig. 4a showed the microstructure of WM near the interface. The microstructure of WM, with nickel content of about 69 wt.%, displayed austenitic columnar dendrites within the

D

solidification grain boundary (SGB). The SGBs are the direct result of competitive growth

PT E

[28].

Fig. 4b showed the grain morphology of WM near the fusion boundary by EBSD. The

CE

SGBs were evident but the solidification subgrain boundaries were not shown in Fig.4b because of its low misorientation (less than five degrees) [29]. In addition, it showed that the

AC

grain growth direction of WM away from the fusion boundary was perpendicular to the fusion boundary because the temperature gradient was largest in the direction perpendicular to the fusion boundary. The columnar grains exhibited different sizes because of competitive growth mechanism. Especially in <100> direction, the columnar dendrites grew more easily than in other direction during non-equilibrium solidification[27]. In order to further study segregation of elements caused by non-equilibrium solidification in WM, the elements distribution were measured by EPMA. The results were shown in Fig. 5 which indicated a strong segregation of Ti and Si in the interdendritic region, while the segregation of Fe was in dendritic core region. Elements such as Cr, Ni and C have no 5

ACCEPTED MANUSCRIPT obvious segregation during solidification. It was noted that Ti and N were enriched in the interdendritic region. The solidification behavior of WM was controlled by solute redistribution behavior. Solute redistribution of alloying elements was related to the diffusivity of solute in the solid (Ds) and the equilibrium distribution coefficient k (k=Cs/C0), where Cs was the solid composition at the solid/liquid interface, C0 was the nominal alloy composition. It was established under the

PT

conditions of equilibrium at the solid/liquid interface, negligible dendrite tip undercooling, and so on. However, the cooling rate had a great influence on the element segregation, another

RI

coefficient K= (Cg/Cb) could be used to characterize the intensity of element segregation,

SC

where Cg was the composition of the interdendritic region, Cb was the composition of the dendrite core. The greater the value of K was, the severer elements segregation became when

NU

k<1. The value of Cg and Cb were quantitatively analyzed by EPMA, and the segregation coefficient of the main elements were calculated and presented in Table 4. It was found that

MA

the K coefficient of Ni and Fe with k>1, was slightly smaller than 1, indicating a slight segregation to dendritic core. As for Cr with k<1, the value of K is higher than 1, therefore its behavior is contrary to elements Ni and Fe. The K coefficients of Ti and Si are much higher

D

than 1, which indicates that Ti and Si segregate to the liquid metal and are rich in the

PT E

interdendritic region at the end of solidification. Fig. 5c and 5d showed that Ti and N were rich in the same position of the cladding. The

CE

further examination of microstructure in this position was carried out by SEM shown in Fig.6a. It showed that there were a large number of cubical precipitates distributed in the

AC

cladding.

Fig. 6b and 6c showed the TEM image, the diffraction pattern and the chemical composition of the precipitates respectively. It can be seen that the precipitate exhibited the cubic morphology with a side length of 500nm.The selected area electron diffraction (SAED) of the precipitate with [001] zone axis showed a face-centered cubic (FCC) crystal structure with a lattice parameter of about 0.432 nm. The EDX result showed that the precipitate contains Ti and N, and the ratio of Ti atoms to N atoms is approximate to one. It confirmed that the precipitate is titanium nitride rather than titanium carbide or carbon nitride. The results of other investigations [30, 31] also proved that the TiN precipitate with fcc structure 6

ACCEPTED MANUSCRIPT was found even though the content of N was extremely low. In addition, when the substrate and the precipitate are the cubic crystal structure in a parallel direction relationship, the precipitate surrounded by the substrate will be cuboidal or spherical morphology [32]. That is the reason that the TiN is cuboidal morphology as shown in Fig. 6b. The formation of the precipitate was attributed to the minor elements and their diffusion, i.e., segregation during solidification. As shown in Table 1, ERNi-1 filler metal contains 2.765

PT

wt.% Ti which can effectively reduce the oxygen content, thus preventing the generation of solidification cracks and pores in Nickel-based alloy. However, it has a strong tendency to

RI

form oxides, nitrides, sulfides, and carbides in sequence [33]. According to the results of

SC

EPMA and TEM, the precipitate in interdenritic region is TiN rather than TiC. The element N in precipitate comes from the melting base metal. The precipitate (TiN) behavior of WM can

NU

be predicted by JMatPro software which is shown in Fig. 6d. The calculation was carried out using the compositions of interdendritic region and the temperature ranged from 800°C to

MA

1600°C. MN was predicted to form from the liquid at 1430.04°C during equilibrium cooling. The composition of MN can be treated as TiN in the liquid due to the extreme low solubility of carbon. With the temperature decreasing, liquid-solid transformation was ongoing and

D

ended at 1399.05 °C. In the range of temperature, the solubility of the nitrogen decreased

PT E

while carbon solubility increased with the decreasing temperature [34]. However, there was not enough time for carbon to diffuse into precipitate because of quick cooling rate. In other

CE

words, the high cooling rate of GTAW led the formation of TiN at the end of solidification. 3.3 Microstructure of the unmixed zone

AC

Fig. 7 showed the microstructure of UZ. It can be found that a large amount of precipitates with vermicular morphology formed with some small dot-like precipitates dispersed among them.

EPMA mapping performed on UZ showed that the precipitates were enriched in element Cr and depleted in element Ni as shown in Fig.8. In order to identify the structure of precipitate, TEM analysis results were shown in Fig. 9. The SAED of the precipitate with [1̅11] zone axis in Fig. 9a showed a body-centered cubic (BCC) crystal structure with a lattice parameter of about 0.286 nm. Fig.9b and 9c showed the composition of matrix and precipitate by EDX respectively. It can be seen that the precipitate 7

ACCEPTED MANUSCRIPT contains 29.69 wt. % Cr which was significantly higher than that in matrix. It was confirmed that the precipitate was δ-ferrite which has a high content of element Cr. The results were consistent with that of studies [35, 36, 37], where reported that δ-ferrite with bcc structure was found in UZ of austenitic stainless steel weldments. In fact, the formation of the UZ was related to convective flow of molten pool. During the welding process, the melted base metal and the filler metal were mixed by intensive

PT

convection in molten pool. However, the unmelted base metal was apparently in a quiescent state. When the molten metal flowed along the solid-liquid interface on the unmelted base

RI

metal, it was necessary to satisfy the non-slip condition of the interface, so that the tangential

SC

flow velocity of liquid metal at the solid-liquid interface was zero. Even though the convection of the fusion pool was very intense, a transition zone appeared where the flow

NU

velocity of molten metal changed from zero to the molten pool flow velocity and convective mass transfer did not occur according to the theory of flowing boundary layer in fluid

MA

mechanics. In other words, the melted base metal in this zone defined as UZ was unmixed into molten pool and re-solidified with elements diffusion [37]. The formation of δ-ferrite in UZ was relative to the ferritic-austenitic (FA) solidification

D

mode of 304SS. In this solidification mode [38], the primary δ-ferrite solidified and

PT E

subsequently the austenite (γ) started to form in the interdendritic regions following primary ferrite through a peritectic-eutectic reaction. When it cooled through the two-phase δ-ferrite and austenite field, δ→γ transformation was controlled by diffusion of elements. The

CE

ferrite-promoting elements (chromium) segregated to δ-ferrite while austenite-promoting

AC

elements (nickel, carbon and nitrogen) diffused to austenite. Finally, the transformation ended at lower temperature and formed vermicular ferrite morphology [39]. 3.4 Microstructure of the heat affected zone (HAZ) Fig. 10a shows the microstructure of BM and HAZ. Some residual high temperature ferrite (δ-ferrite) is paralleled to the rolling direction. Compared with base metal, the grain coarsening effect is evident in HAZ. The grain size of the HAZ and BM were measured by EBSD as shown in Fig. 10b. The average grain diameter of HAZ is 40.5µm, whereas that of BM is 20.2µm according to the results by Tango analysis software. The grain coarsening through the grain boundary migration is influenced by factors such as the peak reheating 8

ACCEPTED MANUSCRIPT temperature, chemical composition, etc. The driving force of grain coarsening is the descent of surface energy, the surface energy of grains in HAZ increased due to their undergone the repeated welding heat cycles. In order to make the system stable, the zigzag grain boundaries with higher surface energy were transformed into flat grain boundaries with lower surface energy by migration. Thus some grains were merged into another grains, thereby it promoted the grain coarsening. In addition, the multi-pass and multi-layer welding further promoted this

PT

phenomenon. Fig.11showed the SEM microstructure of BM and HAZ. It was obvious that no precipitates

RI

were found at the grain boundaries of BM and HAZ. The HAZ undergone high temperature

SC

but its duration was not long enough to form precipitates such as Cr23C6, and Cr2N. 3.5 Hardness distribution

NU

Fig. 12 showed the transverse hardness profile of three cladding layers. The hardness decreased with an increasing distance from BM. The hardness of BM has a value of 218±

MA

8HV.The first cladding layer has the maximum hardness of 179±9HV comparing with the other cladding layers. The hardness values of the second and the third layers are 162±6HV and147±4 HV, respectively. The chemical compositions, especially the content of Fe and Cr,

D

played an important role in determining the hardness of solid-solution strengthened

PT E

Nickel-based alloy. The elements Fe and Cr are solid solution strengtheners, which means the effective of solid-solution hardening becomes better [28] with the content increase of Fe and

CE

Cr. Table. 5 showed the major chemical composition of cladding layers (Note that the reported composition values represent the average of ten individual measurements by EDS). The

AC

contents of Fe and Cr in first cladding layer were higher than that in the second and the third cladding layer as a result of high dilution of substrate during welding. Thus the maximum hardness of cladding layers occurred at the first cladding layer due to the highest content of Fe and Cr [40]. 3.6 Corrosion properties The weight loss of the cladding layer samples immersed in FLiNaK salts at 650℃for200h is about 0.16mg/cm2. Compared with Hastelloy N, the average corrosion rate of the cladding layers is very small which indicates the strong resistance of Ni to molten fluoride salts [18]. Compared with the potentiodynamic polarization curves of Ni, Hastelloy N, and 316ss in 9

ACCEPTED MANUSCRIPT molten FLiNaK salts [21-23], it can be seen that Ni exhibited the best corrosion resistance properties in molten FLiNaK salts at high temperature. Fig. 13 showed the SEM images of the surface and cross-section of cladding layer after the corrosion test. It was obvious that no grain boundaries were attacked by corrosion and no pits can be found below the surface. EPMA mapping performed on the cross-section of the cladding layer showed the

PT

relatively high concentration of Fe in the near-surface region as shown in Fig.14. It was noted that the low concentrations of Fe and Ni occurred at the edge of the sample due to the edge

RI

effect of EPMA measurement. The thickness of Fe-rich layer was 1.5μm which was labeled in

SC

EPMA mapping. Fe2+ impurity in molten salts was firstly deposited on the surface of cladding layer leading the Fe-concentration gradient between the surface and the sample interior.

NU

Secondly, the Fe-concentration gradient promoted the diffusion process of Fe [9]. Many researchers concluded that the corrosion rate in molten fluoride salts was marked by initial

MA

rapid attack associated with dissolution of Cr and was fast driven by the impurities in the salt. In addition, element Ni was hardly dissolved into molten salts [41,42,43]. The Cr content of cladding layer is too low to affect the corrosion rate (As shown in Fig.14c). It can be

PT E

according to the results.

D

confirmed that the cladding layer was virtually immune to be attacked in molten salts

4.Conclusions

CE

(1) ERNi-1 filler metal was feasibly overlaid onto the surface of 304SS using appropriate welding parameters.

AC

(2) The epitaxial growth and columnar dendrites were observed in the cladding near the interface. The precipitate TiN with FCC crystal structure formed in the interdendritic region of WM near the interface because of element segregation during non-equilibrium solidification. (3) The formation of UZ was that a part of BM was melted and then solidified where this BM cannot be mixed into molten pool since the convection current was not able to promote enough fluid flow. The vermicular δ-ferrite was precipitated in UZ. Grain coarsening was evident, and no precipitates except for high temperature δ-ferrite found in HAZ. (4) The hardness of cladding layers decreased with the number of cladding layers due to the 10

ACCEPTED MANUSCRIPT reduction of Fe and Cr content in cladding layers. (5) The cladding layer was fully resistant to the molten salt corrosion. Acknowledgments This work was supported by National key research and development program of China (2016YFB0700404), National Natural Science Foundation of China (Grant No.51501216, 51371188, 51671122, 51671154 and 51601213), and Strategic Priority Research Program of

PT

the Chinese Academy of Sciences (Grant No. XDA02004210) and Talent development fund

AC

CE

PT E

D

MA

NU

SC

RI

of shanghai (201650), and Natural Science Foundation of Shanghai (15ZR1448500).

11

ACCEPTED MANUSCRIPT References [1] J. Serp, M. Allibert, O. Beneš, S. Delpech, O. Feynberg, V. Ghetta, D. Heuer, D. Holcomb, V.V. Ignatiev, J.L. Kloosterman, L. Luzzi, E. Merle-Lucotte, J. Uhlíř, R. Yoshioka, Z.M. Dai, The molten salt reactor (MSR) in generation IV: overview and perspectives, Prog. Nucl. Energy. 77 (2014) 308-319.

PT

[2] P. Rubiolo, D. Heuer, E. Merle-Lucotte, M. Brovchenko, V. Ghetta, M. Allibert, H., A. Laureau, Overview and perspectives of the molten salt fast reactor (MSFR) concept,

RI

Proceedings of the International Conference on Molten Salts in Nuclear Technology

SC

(CMSNT), Mumbai, India, 2013.

[3] S. Delpech, E. Merle-Lucotte, D. Heuer, M. Allibert, V. Ghetta, C. Le-Brun, X. Doligez,

NU

G. Picard,Reactor physics and reprocessing scheme for innovative molten salt reactor

MA

system,J. Fluor. Chem. 130(2009)11-17.

[4] OECD, Nuclear Energy Agency for the Generation International Forum, Technology

D

roadmap update for generation Ⅳnuclear energy systems, Gif-tru2014,January 2014.

PT E

[5] T. Liu, J.S. Dong, L. Wang, Z.J. Li, X.T. Zhou, L.H. Lou, J. Zhang, Effect of long-term thermal exposure on microstructure and stress rupture properties of GH3535 superalloy, J.

CE

Mater. Sci. Tec. 31 (2015) 269-279. [6] F.F. Han, B.M. Zhou, H.F. Huang, B. Leng, Y.L. Lu, Z.J. Li, X.T. Zhou, Effect of

AC

long-term thermal exposure on the hot ductility behavior of GH3535 alloy,Mater. Sci. Eng., A. 673 (2016) 299-306. [7] F.F. Han, B.M. Zhou, H.F. Huang, B. Leng, Y.L. Lu, J.S. Dong, Z.J. Li, X.T. Zhou, The tensile behavior of GH3535 superalloy at elevated temperature, Mater. Chem. Phys. 182 (2016) 22-31. [8] X.X. Ye, H. Ai, Z. Guo, H.F. Huang, L. Jiang, J.Q. Wang，Z.J. Li，X.T. Zhou，The high-temperature corrosion of Hastelloy N alloy (UNS N10003) in molten fluoride salts analysed by STXM, XAS, XRD, SEM, EPMA, TEM/EDS,Corros. Sci. 106 (2016) 12

ACCEPTED MANUSCRIPT 249-259. [9] ASME,ASME Boiler and Pressure Vessel Code Division1-Subsection NH, New York, 2015, pp. 21-24. [10] M. Brandt, S. Sun, N. Alam, P. Bendeich, A. Bishop, Laser cladding repair of turbine

PT

blades in power plants: From research to commercialization, IHTSE. 3 (2013) 105-114. [11] E. Capello, D. Colombo, B. Previtali, Repairing of sintered tools using laser cladding by

RI

wire, J. Mater. Process. Tech. 164-165 (2005) 990-1000.

SC

[12] C. Wang, M. Zhu, X.Z. Wang, Y. Wang, Study on overlay welding of dissimilar steel by consumable double electrode gas metal arc welding method with one power supply, Hot

NU

Working Technology. 45 (2016) 180-182.

MA

[13] F. Gillemot, Overview of reactor pressure vessel cladding, Int. J. Nuclear Knowledge Management. 4 (2010) 265-278.

SA508–309L/308L–316L

PT E

an

D

[14] H.L. Ming, Z.M. Zhang, J.Q. Wang, E.H. Han, W. Ke, Microstructural characterization of domestic

dissimilar

metal

welded

safe-end

joint,Mater.Charact. 97 (2014) 101–115.

CE

[15] J.N. DuPont, S.W. Banovic, A.R. Marder, Microstructural evolution and weldability of dissimilar welds between a super austenitic stainless steel and nickel base alloys, Weld. J.

AC

82 (2003) 125-135.

[16] A. Mortezaie, M. Shamanian, An assessment of microstructure, mechanical properties and corrosionresistance of dissimilar welds between Inconel 718 and 310S austenitic stainless steel,Int. J. Pres. Ves. Pip. 116 (2014) 37-46. [17] J. Beddoes, J.G.Parr, Introduction to stainless steels, third ed., ASM International, Materials Park, OH, USA, 1999. [18] L.C. Olson, J.W. Ambrosek, K. Sridharan, M.H. Anderson, T.R. Allen, Materials 13

ACCEPTED MANUSCRIPT corrosion in molten LiF–NaF–KF salt, J. Fluor. Chem. 130 (2009) 67–73. [19] Y.L. Wang, H.J. Liu, C.L. Zeng, Galvanic corrosion of pure metals in molten fluorides, J. Fluor. Chem. 165 (2014) 1-6. [20] C.S. Sona, B.D. Gajbhiye, P.V. Hule, A.W. Patwardhan, C.S. Mathpati, A. Borgohain, N.K. Maheshwari, High temperature corrosion studies in molten salt-FLiNaK, Corros.

PT

Eng. Sci. Technol. 49 (2014) 287-295.

RI

[21] Y.L. Wang, H.J. Liu, G.J. Yu, J. Hou, C.L. Zeng, Electrochemical study of the corrosion

SC

of a Ni-based alloy GH3535in molten (Li,Na,K)F at 700,J. Fluor. Chem.178(2015)14-22. [22] Y.L. Wang, Q. Wang, H.J. Liu, C.L. Zeng, Effects of the oxidants H2O and CrF3on the

NU

corrosion of pure metals inmolten (Li,Na,K)F, Corros. Sci. 103 (2016) 268-282.

MA

[23] X.B. Ding, Investigation on the corrosion behavior of Hastelloy N and 316L stainless steel in LiF-NaF-KF molten salt. The university of Chinese Academy of Sciences,

D

Shanghai, China,2015,pp.31-35.

(1995) 406-416.

PT E

[24] J.N. Dupont, A.R. Marder, Thermal efficiency of arc welding processes, Weld. J. 74

CE

[25] A. Strecker, U. Salzberger, J. Mayer, Specimen preparation for transmission electron microscopy: reliable method for cross-sections and brittle materials, Prakt. Metallogr. 30

AC

(1993) 482.

[26] R.S. Sellers, W.J. Cheng, B.C. Kelleher, M.H. Anderson, K. Sridharan, C.J. Wang, T.R. Allen, Corrosion of 316L stainless steel alloy and Hastelloy-N superalloy in molten eutectic LiF-NaF-KF salt and interaction with graphite, Nucl. Technol. 188 (2014) 185-191. [27] S. Kou, Welding Metallurgy, third ed., John Wiley&Sons, Inc., New Jersey, 2003. [28] J.N. Dupont, J.C. Lippod, S.D. Kiser, Welding metallurgy and weldability of Nickel-base 14

ACCEPTED MANUSCRIPT alloys, John Wiley&Sons, Inc., New Jersey, 2009. [29] J.C. Lippod, W.A.T. Clark, M. Tumuluru, An investigation of weld metal interfacesin: M.J. Cieslak, J.H. Perepezlo, M.E. Glicksman (Eds.), The metal science of joining, The Minerals, Metals & Materials Society, Warrendale (PA), 1992, pp. 141-146. The Metal Science of Joining. (1992) 141-146.

PT

[30] M. Joonoh, L.Changhee, U. Sangho, L. Jongbong, Coarsening kinetics of TiN particle in a low alloyed steel in weld HAZ: considering critical particle size, Acta. Mater. 54 (2006)

SC

RI

1053–1061.

[31] H.H. Jin, J.H. Shim, Y.W. Cho, H.C. Lee, Formation of intragranular acicular ferrite

NU

grainsin a Ti-containing low carbon steel, High Temp. Mater. Proc. 34 (2015) 1111-1113. [32] M. Enomoto, Z.G. Yang, T. Nagano, Calculation of the equilibrium shape of TiN particles

MA

in iron, ISIJ Int. 44 (2004) 1454–1456.

D

[33] R. Mendoza, J. Camacho, G. Lugo, C. Lopez, L. Herrera, J. Reyes, C. Gonalez, J.A.

PT E

Juarez-Islas, Structure of a low carbon Al-killed/Ti-added steel, ISIJ Int. 37 (1997) 176-180.

CE

[34] L.F.S. Dumitrescu, M. Hillert, Reassessment of the solubility of TiC and TiN in Fe, ISIJ. Int. 39 (1999) 84-90.

AC

[35] A. Ziewiec, E. Tasak, A. Zielinska, K. Ziewiec, J. Kowalska, The influence of rapid solidification on the microstructure of the 17Cr–9Ni–3Mo precipitation hardened steel, J. Alloys Compd. 615 (2014) 627-632. [36] D.M.E. Villanueva, F.C.P. Junior, R.L. Plaut, A.F. Padilha, Comparative study on sigma phase precipitation of three types of stainless steels: austenitic, superferritic and duplex, Mater. Sci. Tech-Lond. 22(2006) 1098-1104. [37] W.A. Baeslack, J.C. Lipplod, W.F. Savage, Unmixed zone formation in austenitic 15

ACCEPTED MANUSCRIPT stainless steel weldment, Weld. J. 58 (1979) 168-176. [38] J.A. Brooks, A. W. Thompson, Microstructural development and solidification cracking susceptibility of austenitic stainless steel welds, Int. Mater. Rev. 36 (1991) 16-44. [39] A. Hunter, M.Ferry, Phase formation during solidification of AISI 304austenitic stainless

PT

steel, Scripta. Mater. 46 (2002) 253–258. [40] C.M. Lin, T.L. Su, K.Y, Wu, Effects of parameter optimization on microstructure and

RI

properties of GTAW clad welding on AISI 304L stainless steel using Inconel 52M, Int. J.

SC

Adv. Manuf. Technol. 79 (2015) 2057–2066.

[41] ORNL, in: R.B. Evans III, J.H. Devan,G.MWatson, Self-diffusion of chromium in

NU

nickel-base alloys, ORNL-2982, Oak Ridge, TN, USA, 1960.

MA

[42] M. Kondo, T. Nagasaka, A. Sagara, T. Muroga, D. Oyama, M. Suzuki, T. Terai, Corrosion characteristics of reduced activation ferritic steel JLF-1 (8.92Cr-2W) in molten salts Flibe

D

and Flinak, Fus. Eng. Des. 84 (2009) 1081–1085.

PT E

[43] P.Calderoni, C.Cabet, Corrosion issues in molten salt reactor (MSR) systems in: DanienFeron (Eds.), Nuclear corrosion science and engineering, Woohead publishing

AC

CE

series in Energy, 2012, pp. 842-863.

16

ACCEPTED MANUSCRIPT

Table 1. Chemical composition of the base metal and filler metal (wt.%). Mn

Ni

Cr

Ti

Al

Fe

P

S

C

N

Cu

304SS

0.4

1.18

8.02

18.14

-

-

Bal.

0.028

0.001

0.044

0.048

-

ERNi-1

0.363

0.501

95.64

-

2.765

0.409

0.069

0.006

0.005

0.03

-

0.11

AC

CE

PT E

D

MA

NU

SC

RI

PT

Si

17

ACCEPTED MANUSCRIPT

Table 2.The GTAW welding parameters. Filler

Current

Voltage

welding speed

feed rate

heat input

arc swing

material

(A)

(V)

(mm/min)

(m/min)

(kJ/mm)

(mm)

ERNi-1

250

11

100

1.8

AC

CE

PT E

D

MA

NU

SC

RI

PT

1.15

18

6

ACCEPTED MANUSCRIPT

Table 3. Chemical composition of cladding near the interface (wt%). Fe

Cr

Ti

Al

Si

69.24

22.44

5.51

1.97

0.47

0.36

AC

CE

PT E

D

MA

NU

SC

RI

PT

Ni

19

ACCEPTED MANUSCRIPT

Table 4.Chemical composition in interdendritic region (Cg), the dendrite core (Cb) and

Cg

Cb

K

Ni

69.41

70.71

0.98

Fe

20.17

22

Cr

5.28

5.19

Ti

4.4

1.73

2.54

Si

0.74

1.94

RI

PT

Element

SC

segregation coefficient K.

AC

CE

PT E

D

MA

NU

0.38

20

0.92 1.02

ACCEPTED MANUSCRIPT

Table 5.Chemical composition of cladding layers Clad 1 layer

Clad 2 layer

Clad 3 layer

Ni

72.23±5.19

89.5±2.1

94.75±0.35

Fe

19.12±3.88

5.87±1.59

Cr

4.61±1.22

1.38±0.45

AC

CE

PT E

D

MA

NU

SC

RI

PT

Elements

21

1.59±0.24 0.38±0.11

ACCEPTED MANUSCRIPT Figure captions: Fig. 1. Schematic diagram of corrosion experimental setup. Fig. 2.Optical micrograph of claddinginterface (a) and back scattered electron image of interface between UZ and WM (b).

PT

Fig. 3. Elements distributionacross the interface by EDS.

RI

Fig. 4.Optical micrograph of WM (a) and EBSD micrograph of WM (b).

SC

Fig. 5.Back scattering electron image (a) and EPMA mapping showing relative concentration for the dendritic core and interdendritic region, (b) C, (c) N, (d) Ti, (e) Si, (f) Fe, (g) Cr, (h)

NU

Ni.

MA

Fig. 6. SEM micrographof the precipitate in WM (a),TEM micrograph with its SAED pattern of the precipitate (b), chemical composition of the precipitate (c), Element Ti, N, Cdistribution in the precipitate (MN, the M denotes Ti) during equilibrium cooling at the

PT E

D

temperature 800°Cto 1600°C(d).

Fig. 7. SEM micrograph of the UZ microstructure (a) and magnified micrograph (b).

CE

Fig. 8. Back scattering electron image (a) and EPMA mapping of elements (b) Cr, (c) Fe, (d)

AC

Ni, (e) C, (f) N.

Fig. 9. TEM micrograph of the precipitate and its SAED pattern in UZ (a), the chemical composition of the precipitate (b) and the matrix in UZ (c). Fig. 10.Optical micrograph of BM and HAZ(a) and EBSD micrograph of BM and HAZ (b). Fig. 11. SEM micrographs of HAZ (a) and BM (b). Fig. 12. Hardness distribution from 304SS substrate to cladding layers. Fig. 13. SEM images of cladding layer after corrosion test in molten FLiNaK at 650 ºC for 22

ACCEPTED MANUSCRIPT 200h. Surface (a) and cross section (b). Fig. 14.Cross-section of back scattering electron images (a) and EPMA mapping of the cladding layer after corrosion testing in molten FLiNaK salts at 650 ºC for 200h, showing the

AC

CE

PT E

D

MA

NU

SC

RI

PT

elemental distribution of Fe(b), Ni (c), Cr (d), Ti (e).

23

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Fig. 1

24

AC

CE

PT E

D

MA

NU

Fig. 2

SC

RI

PT

ACCEPTED MANUSCRIPT

25

Fig. 3

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

26

AC

CE

PT E

D

MA

NU

Fig. 4

SC

RI

PT

ACCEPTED MANUSCRIPT

27

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

Fig. 5

28

Fig. 6

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

29

AC

CE

PT E

D

MA

NU

Fig. 7

SC

RI

PT

ACCEPTED MANUSCRIPT

30

Fig. 8

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

31

Fig. 9

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

32

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

Fig. 10

33

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

Fig. 11

34

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

Fig. 12

35

AC

CE

PT E

D

MA

NU

SC

Fig. 13

RI

PT

ACCEPTED MANUSCRIPT

36

Fig. 14

AC

CE

PT E

D

MA

NU

SC

RI

PT

ACCEPTED MANUSCRIPT

37

ACCEPTED MANUSCRIPT

AC

CE

PT E

D

MA

NU

SC

RI

PT

Graphical abstract

38

ACCEPTED MANUSCRIPT Highlights 

Ni-201 alloy was successfully deposited onto 304SS for the first time for molten salt reactor. The microstructure of interface was characterized by a series of methods.



The formation of TiN in weld metal was investigated.

AC

CE

PT E

D

MA

NU

SC

RI

PT



39